Values and Uncertainties in the Predictions of Global Climate Models

Abstract

Over the last several years, there has been an explosion of interest and attention devoted to the problem of Uncertainty Quantification (UQ) in climate science—that is, to giving quantitative estimates of the degree of uncertainty associated with the predictions of global and regional climate models. The technical challenges associated with this project are formidable, and so the statistical community has understandably devoted itself primarily to overcoming them. But even as these technical challenges are being met, a number of persistent conceptual difficulties remain. So why is UQ so important in climate science? UQ, I would like to argue, is first and foremost a tool for communicating knowledge from experts to policy makers in a way that is meant to be free from the influence of social and ethical values. But the standard ways of using probabilities to separate ethical and social values from scientific practice cannot be applied in a great deal of climate modeling, because the roles of values in creating the models cannot be discerned after the fact—the models are too complex and the result of too much distributed epistemic labor. I argue, therefore, that typical approaches for handling ethical/social values in science do not work well here.

Ontological Choices and the Value-Free Ideal

“…Given that recent debates about science and values tend to focus on issues of uncertainty and induction (e.g. Winsberg 2012;Biddle 2013;Morrison 2014;Intemann 2015;Leuschner 2015cf. Brown 2013, a discussion of ontological choices broadens the perspective of debates about social values in scientific practice.…”

Practice and philosophy of climate model tuning across six US modeling centers

“…It is these additional degrees of freedom that are used to "tune" or calibrate the emergent properties of the model against a selected set of target observations. The decisions on what to tune, and especially what targets to tune for, undoubtedly involve value judgments (Hourdin et al, 2017;Winsberg, 2012;Schmidt and Sherwood, 2014;Intemann, 2015). Notably, there is not any obvious consensus in the modeling community as to the extent to which parameter choices should be guided by conforming to process-level knowledge as opposed to optimizing emergent behaviors in climate models.…”

Values and evidence: how models make a difference

“…The selection of purposes and priorities in turn can influence decisions in model construction and evaluation (Winsberg 2012;Elliott and McKaughan 2014;Intemann 2015). When it comes to model construction, for instance, purposes and priorities shape not just which entities and processes are represented in a model but also how they are represented, including which simplifications, idealizations and other distortions are more or less tolerable.…”

“…9 This is the situation when it comes to the weather forecasting models discussed in the last section and even more so for their cousins, global climate models (see also Biddle and Winsberg 2009;Winsberg 2012). In practice, it is often infeasible to adjust model-based estimates to arrive at confident and precise assignments for p(H|e&B).…”

“…Our argument, which builds upon but generalizes beyond recent work focusing on values in climate modeling (Biddle and Winsberg 2009;Winsberg 2012;Parker 2014;Intemann 2015), focuses on the use of scientific models -especially mathematical models and computer simulation models -as surrogates for background knowledge in Bayesian and other studies that aim to assign probabilities to scientific hypotheses. Because these scientific models often deviate from background knowledge in ways that are in part dependent on non-epistemic values, the probabilities estimated via these studies also turn out to be dependent on non-epistemic values in the sense that, if the non-epistemic values had been different, so would have the estimated probabilities, even with the same background knowledge.…”